Successful protein folding involves the conversion of a linear polypeptide into a stable and biologically active 3D structure. This process has been studied for decades by denaturing purified, full-length polypeptides in vitro, diluting away the denaturant, and observing the refolding process. What is still lacking, however, is an understanding of how these in vitro results relate to protein folding in vivo. The initiation of protein folding in vivo is fundamentally different: the starting ensemble for folding in vivo is a growing nascent polypeptide chain, rather than a full-length chain. We have very little information about how growth of the nascent chain affects the energy landscape for folding. Yet this difference may help explain why some native state topologies are well represented in vivo, but challenging to refold in vitro as full-length polypeptide chains. In this proposal, the influence of translation on protein folding will be investigated from three unique perspectives: (1) Early events: nascent chain diffusion and collapse. Our collaborator Lisa Lapidus has recently shown that intramolecular diffusion in vitro is extremely slow for an unfolded, collapsed protein in buffer, relative to high concentrations of denaturant. We have adapted Lapidus's methods to determine the extent to which a ribosome-bound nascent chain can collapse, and the intramolecular diffusion rate for that collapsed state, as these parameters will affect the size and shape of the energy landscape accessible to that nascent chain after its release from the ribosome. (2) Synonymous rare codons reduce local translation rate, which can alter co-translational folding mechanisms, presumably by altering the accessible energy landscape for folding. Remarkably, recent results emerging from our lab and others now indicate that, for some proteins, local translation rate can also alter the native structure, enabling access to an energy minimum not accessible under other translation rate patterns. We will measure the capacity of synonymous codon selection (mRNA sequence) to modify protein structure, thereby characterizing its modifications to the energy landscape for folding in vivo. (3) In the first project period, we showed that synonymous rare codons are not randomly distributed along gene sequences, but tend to cluster together, despite established negative effects of such clustering (including reduced translation rate). What positive cellular effects might overcome these known negatives? Certainly modified co-translational folding (to increase the yield of correctly folded protein) is one effect, but broader effects, including impact on other cell functions, must also be considered. We have developed a tunable in vivo system with which to answer previously unanswerable questions regarding the effects of co-translational folding on cell physiology, including cell fitness. Taken together, results from this proposal will reveal how translation modifies early folding events, native protein structures, and cell physiology. Several decades of in vitro refolding studies have revealed general principles for protein refolding in the test tube; results from this proposal will be used to develop principles for protein folding in vivo.